Skip to content

Borrowing gaming DQN agents for eventual appliction in Financial data

Notifications You must be signed in to change notification settings

kimardenmiller/Deep-QNetwork_gaming

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Deep Reinforcement Learning from Games to Value Investing

What is it?

State-of-the art deep reinforcement learning algorithm in Tensorflow and Python for eventual application to non-gaming environments.

The performance above came after:

  • Playing 1.75 million games over
  • ~ 2 days of training
  • 3 x 2D ConvNet, and
  • a Linear Annealed Greedy EPS Policy

Todos

  • Convert agent to work on MktCap / GVA application

About

Borrowing gaming DQN agents for eventual appliction in Financial data

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages